Untitled
Baiheng Chen
2023/4/8
Read in the data
This is a classical dataset about Breast cancer from Kaggle. https://www.kaggle.com/datasets/uciml/breast-cancer-wisconsin-data In some pervious reasearch, some machine learning methods are used to do the binary classification. The dataset here contains 10 dimensions clincal features
library(readr)
library(ggplot2)
library(dplyr)## Warning: 程辑包'dplyr'是用R版本4.2.3 来建造的##
## 载入程辑包：'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2
## ──## ✔ tibble 3.2.1 ✔ stringr 1.5.0
## ✔ tidyr 1.2.1 ✔ forcats 1.0.0
## ✔ purrr 1.0.0## Warning: 程辑包'tibble'是用R版本4.2.3 来建造的## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(tidymodels)## Warning: 程辑包'tidymodels'是用R版本4.2.3 来建造的## ── Attaching packages ────────────────────────────────────── tidymodels 1.0.0 ──
## ✔ broom 1.0.3 ✔ rsample 1.1.1
## ✔ dials 1.1.0 ✔ tune 1.0.1
## ✔ infer 1.0.4 ✔ workflows 1.1.3
## ✔ modeldata 1.1.0 ✔ workflowsets 1.0.0
## ✔ parsnip 1.0.4 ✔ yardstick 1.1.0
## ✔ recipes 1.0.5## Warning: 程辑包'dials'是用R版本4.2.3 来建造的## Warning: 程辑包'infer'是用R版本4.2.3 来建造的## Warning: 程辑包'modeldata'是用R版本4.2.3 来建造的## Warning: 程辑包'parsnip'是用R版本4.2.3 来建造的## Warning: 程辑包'rsample'是用R版本4.2.3 来建造的## Warning: 程辑包'tune'是用R版本4.2.3 来建造的## Warning: 程辑包'workflows'是用R版本4.2.3 来建造的## Warning: 程辑包'workflowsets'是用R版本4.2.3 来建造的## Warning: 程辑包'yardstick'是用R版本4.2.3 来建造的## ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
## ✖ scales::discard() masks purrr::discard()
## ✖ dplyr::filter() masks stats::filter()
## ✖ recipes::fixed() masks stringr::fixed()
## ✖ dplyr::lag() masks stats::lag()
## ✖ yardstick::spec() masks readr::spec()
## ✖ recipes::step() masks stats::step()
## • Learn how to get started at https://www.tidymodels.org/start/library(patchwork)
library(embed)## Warning: 程辑包'embed'是用R版本4.2.3 来建造的url <- "https://raw.githubusercontent.com/900Step/Stat-436/main/cancer.csv"
cancer <- read_csv(url)## Rows: 569 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): diagnosis
## dbl (11): id, radius_mean, texture_mean, perimeter_mean, area_mean, smoothne...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.# cancer
pca_result <- prcomp(cancer[,3:12], center = TRUE, scale. = TRUE)
# pc_weights
# summary(pc_weights)
pca_summary <- summary(pca_result)
pca_summary ## Importance of components:
## PC1 PC2 PC3 PC4 PC5 PC6 PC7
## Standard deviation 2.3406 1.5870 0.93841 0.7064 0.61036 0.35234 0.28299
## Proportion of Variance 0.5479 0.2519 0.08806 0.0499 0.03725 0.01241 0.00801
## Cumulative Proportion 0.5479 0.7997 0.88779 0.9377 0.97495 0.98736 0.99537
## PC8 PC9 PC10
## Standard deviation 0.18679 0.10552 0.01680
## Proportion of Variance 0.00349 0.00111 0.00003
## Cumulative Proportion 0.99886 0.99997 1.00000# prop_var <- pca_summary$sdev^2 / sum(pca_summary$sdev^2)
# prop_var
# sum(prop_var[1:5])/sum(prop_var)diagnosis = cancer[,c(1,2)]
pca_recipe <- recipe(~., data = cancer) %>%
# update_role(id, starts_with("group"), new_role = "id") %>%
update_role(id, diagnosis, new_role = "id") %>%
step_normalize(all_predictors()) %>%
step_pca(all_predictors())
pca_recipe## ## ── Recipe ──────────────────────────────────────────────────────────────────────## ## ── Inputs## Number of variables by role## predictor: 10
## id: 2## ## ── Operations## • Centering and scaling for: all_predictors()## • PCA extraction with: all_predictors()pca_prep <- prep(pca_recipe)pca_result <- tidy(pca_prep, 2)
pca_result %>% filter(component %in% str_c("PC", 1:5))## # A tibble: 50 × 4
## terms value component id
## <chr> <dbl> <chr> <chr>
## 1 radius_mean -0.364 PC1 pca_Kj431
## 2 texture_mean -0.154 PC1 pca_Kj431
## 3 perimeter_mean -0.376 PC1 pca_Kj431
## 4 area_mean -0.364 PC1 pca_Kj431
## 5 smoothness_mean -0.232 PC1 pca_Kj431
## 6 compactness_mean -0.364 PC1 pca_Kj431
## 7 concavity_mean -0.396 PC1 pca_Kj431
## 8 concave_points_mean -0.418 PC1 pca_Kj431
## 9 symmetry_mean -0.215 PC1 pca_Kj431
## 10 fractal_dimension_mean -0.0718 PC1 pca_Kj431
## # … with 40 more rows# pca_result %>% filter(colnames(pca_result$rotation) %in% str_c("PC", 1:5))
# pca_result
ggplot(pca_result %>% filter(component %in% str_c("PC", 1:5))) +
geom_col(aes(x = value, y = terms)) +
facet_wrap(~ component) +
labs(x = "Component", y = "Features")
pca_scores <- bake(pca_prep, cancer)
# pca_scoresggplot(pca_scores)+
geom_point(aes(x = PC1, y = PC2, color = diagnosis))
group_order <- pca_scores %>%
group_by(diagnosis) %>%
summarise(mpc2 = mean(PC2)) %>%
arrange(mpc2)
group_order## # A tibble: 2 × 2
## diagnosis mpc2
## <fct> <dbl>
## 1 B -0.171
## 2 M 0.288# pca_scores %>%
# mutate(group = factor(diagnosis, levels = group_order)) %>%
# ggplot(aes(x = PC1, y = PC2)) +
# geom_vline(xintercept = 0, col = "#4a4a4a") +
# geom_hline(yintercept = 0, col = "#4a4a4a") +
# geom_point(size = 0.4, alpha = 0.6) +
# scale_x_continuous(breaks = seq(-8, 0, length.out = 3)) +
# scale_color_brewer(palette = "Set2") +
# facet_wrap(~ reorder(group, PC1), ncol = 9) +
# coord_fixed() +
# theme(strip.text = element_text(size = 8))
library(plotly)##
## 载入程辑包：'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layout# Create some sample data
# Create a 3D scatter plot
fig <- plot_ly(x = pca_scores$PC1, y = pca_scores$PC2, z = pca_scores$PC3, type = "scatter3d", mode = "markers", color = pca_scores$diagnosis)
fig## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levelsUMAP
# create a shiny app for the n_neighbor
umap_rec <- recipe(~., data = cancer) %>%
update_role(id, diagnosis, new_role = "id") %>%
step_umap(all_predictors(), learn_rate = 0.1, neighbors = 20)
umap_prep <- prep(umap_rec)
umap_prep ## ## ── Recipe ──────────────────────────────────────────────────────────────────────## ## ── Inputs## Number of variables by role## predictor: 10
## id: 2## ## ── Training information## Training data contained 569 data points and no incomplete rows.## ## ── Operations## • UMAP embedding for: radius_mean, texture_mean, perimeter_mean, ... | Trainedembeddings <- bake(umap_prep, new_data = cancer) %>%
left_join(cancer)## Joining with `by = join_by(id, diagnosis)`ggplot(embeddings) +
geom_point(aes(UMAP1, UMAP2, group = diagnosis, col = diagnosis), alpha = 0.4) +
scale_color_brewer(palette = "Set2")
# embeddings$diagnosis
# , col = as.factor(diagnosis)library(shiny)##
## 载入程辑包：'shiny'## The following object is masked from 'package:infer':
##
## observeui<-fluidPage(
titlePanel("The UMAP plot of the breast cance dataset"),
numericInput("input_n", "n_neighbors: ", value = 50, min = 5, max = 200),
plotOutput("umap")
)
server <- function(input, output){
umap_prep <- reactive({
recipe(~., data = cancer) %>%
update_role(id, diagnosis, new_role = "id") %>%
step_umap(all_predictors(), learn_rate = 0.1, neighbors = input$input_n)%>%
prep()
})
umap_embeddings <- reactive({
bake(umap_prep(), new_data = cancer) %>%
left_join(cancer)
})
output$umap <- renderPlot({
ggplot(umap_embeddings())+
geom_point(aes(UMAP1, UMAP2, group = diagnosis, col = diagnosis), alpha = 0.4) +
scale_color_brewer(palette = "Set2")
})
}
# shinyApp(ui, server)